Let's Make Robots!

Robotic Defense: Brainstorming for future thesis

I'm back after a long summer of work, play, and vacation. Anyhow, I have my foot in the door at a local university and wish to start a thesis early on in my career as an engineering student. Nothing too advanced, but wish to aim for something specific and in depth over the defensive aspect of robots. As in robots to protect yourself or robots to protect themselves.  Right now my ideas are: environmental awareness, protection of circuits, recognizing injury, adaptation to overcome that injury, ect. please let me know which one you think is best, and feel free to challenge my ideas!

 

 

 

 

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

I'm guessing a net of criss-crossing wires can cover the outer shell of the robot, this way it can know where it's hurt if some of them stop conducting. Then, again, the robot still needs to repair itself...

thanks!  good thoughts!

Also periodic sanity checks would be a nice addition. From time to time, the robot can verify that everything is owrking fine i.e. driving forward and seeing if the motors work, putting its hand in front of the camera to see if it detects it, putting it in front of its sonars, etc.

For adaptation to injury, well there's not much that I can think of, but protecting the core is a must. Then, if the core is solid enough, only the wires will break. Here the robot can have lets say three wires going by different places for the same piece, in case any gets damaged. Also some sort of AI would handle the walking gaits and try to find new ones if it loses a leg.

Not sure if this helps. If a robot can measure current drawn by a motor without load (the equivalent of stretching muscles after a good nights sleep) and compares it with figures stored in EEPROM memory then a significant increase in current draw from a motor due to wear or exesive friction could be recognized as pain.
Not sure if this helps. If a robot can measure current drawn by a motor without load (the equivalent of stretching muscles after a good nights sleep) and compares it with figures stored in EEPROM memory then a significant increase in current draw from a motor due to wear or exesive friction could be recognized as pain.

Thanks! Good idea as well!!!

or even wiring. wire are like our major arteries... severe one and your dead. so, adapting to a broken wire even.